Face recognition based on Volterra kernels direct discriminant analysis and effective feature classification

Feng, Guang; Li, Hengjian; Dong, Jiwen; Zhang, Jiashu

doi:10.1016/j.ins.2018.02.028

Cited by 12 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, partition the pixel pairs into two sub-sets; short distance pairs 1 and long distance pairs 2 that is mathematically indicated in the Eqs. (9) and (10).…”

Section: Binary Robust Invariant Scalable Key Pointsmentioning

confidence: 99%

“…In addition, the dimensionality of collected human face images is often very high, which leads to high computational complexity and a curse of dimensionality issue [7,8]. Currently, various dimensionality reduction techniques are developed in semi-supervised, unsupervised, and supervised circumstances like Volterra kernels direct discriminant analysis [9], principal component analysis [10], linear discriminant analysis [10], etc. Additionally, the prior approaches classify the individuals on the basis of hand designed manner that consumes more time for individual classification [11,12].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Face Recognition Using Enhanced Firefly Optimization Algorithm and Deep Belief Network

Prakash¹

2020

IJIES

View full text Add to dashboard Cite

Face recognition is an emerging research area in biometric identification systems, which attracted many researchers in pattern recognition and computer vision. Since, the human facial images are high dimensional in nature, so a new optimization algorithm was proposed in this paper to reduce the dimension of facial images. At first, the input facial images were collected from ORL, YALE, and FAce Semantic SEGmentation (FASSEG) databases. Then, feature extraction was then performed by using Local Ternary Pattern (LTP) and Binary Robust Invariant Scalable Key points (BRISK) to extract the features from the face images. In addition, an enhanced fire-fly optimizer was used to reject the irrelevant features or to reduce the dimension of the extracted features. In enhanced fire-fly optimization algorithm, Chi square distance measure was used to find the distance between the fire-flies and also it uses only a limited number of feature values for representing the data that effectively reduces the "curse of dimensionality" concern. Finally, Deep Belief Network (DBN) was used to classify the individual's facial images. The experimental outcome showed that the proposed work improved recognition accuracy up to 7-20% compared to the existing work.

show abstract

“…Then, partition the pixel pairs into two sub-sets; short distance pairs 1 and long distance pairs 2 that is mathematically indicated in the Eqs. (9) and (10).…”

Section: Binary Robust Invariant Scalable Key Pointsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automatic Face Recognition Using Enhanced Firefly Optimization Algorithm and Deep Belief Network

Prakash¹

2020

IJIES

View full text Add to dashboard Cite

show abstract

“…Non‐linear discrete‐time systems can be frequently modelled by Volterra series [15–18], which is a stable functional series expansion of a non‐linear time‐invariant system. Such series take into account memory effects, which cannot be performed with static non‐linear models [19, 20]. Considering

N \in double-struckN

as the memory depth,

{\overset{false¯}{w}}_{p}^{⋆} false(n_{1}, \dots, n_{p} false) \in double-struckR

as the coefficients of the p th order polynomial basis function and

P \in double-struckN

as the maximum non‐linearity order, in this Letter the reference signal

d false(k false)

is modelled as the sum of three components

\begin{matrix} right leftthickmathspace.5em \\ d (k) & = {()(, w^{⋆})}^{T} bold-italicx (k) + ν_{M} (k) \\ + \overset{≜ ν_{NLC} (k)}{\overset{⏞}{\sum_{p = 2}^{P} \sum_{n_{1} = 0}^{N - 1} \dots \sum_{n_{P} = 0}^{N - 1} {\overset{false¯}{w}}_{p}^{⋆} false(n_{1}, \dots, n_{P} false) \prod_{l = 1}^{p} x false(n - n_{l} false)}}, \end{matrix}

where the first right‐side term corresponds to the affine‐in‐the‐parameters linear component (LC, with

{bold-italicw}^{⋆} \in R^{N}

denoting the unknown and ideal coefficients of the LC),

ν_{normalM} false(k false)

is the measurement noise and

…”

Section: Non‐linear System Modelmentioning

confidence: 99%

“…Non-linear system model: Non-linear discrete-time systems can be frequently modelled by Volterra series [15][16][17][18], which is a stable functional series expansion of a non-linear time-invariant system. Such series take into account memory effects, which cannot be performed with static non-linear models [19,20]…”

mentioning

confidence: 99%

Exact expectation analysis of the LMS adaptive identification of non‐linear systems

et al. 2020

View full text Add to dashboard Cite

Most adaptive filtering schemes employ the tapped-delay line. In part, such a fact can be explained by the assumption that the plant they intend to estimate is linear. Although such a hypothesis can be reasonable if the input signal is constrained to a certain range, sometimes it may not be valid. In this last case, the performance and stability guarantees provided by stochastic models that presume linearity of the ideal system are no longer valid. This Letter advances an analytic model of the least-mean-square (LMS) learning capabilities when the ideal system is not linear, with the additive noise including nonlinear functions of the input samples. The proposed analysis does not assume neither that the excitation signal is statistically independent of the adaptive weights, nor that the additive noise is white and/or independent of input data. Furthermore, it can be applied to non-Gaussian and/or non-white input signals. Simulations show that the advanced model is more accurate than traditional approaches. CAPES, CNPq, and FAPERJ for supporting this research.

show abstract

“…And the local binary patterns are extracted from the Weber-face images which further alleviate the effect of varying illumination. Feng et al [21] presented a novel face recognition method based on direct discriminant Volterra kernels and effective feature classification (DD-VK). This method can simultaneously maximize inter-class distances and minimize intra-class distances in the feature space.…”

Section: Introductionmentioning

confidence: 99%

An Improved Two-Step Face Recognition Algorithm Based on Sparse Representation

et al. 2019

View full text Add to dashboard Cite

Weighted score fusion is a widely used score fusion scheme, but the weights need to be set manually. The results generally vary greatly when the weights are different, so it is difficult to find the optimal weights. This is why it is necessary to constantly set different weights for experimental comparisons to find the optimal weights. In this paper, an improved fusion method is proposed for above shortcoming, that is, multiplication fusion applicable to sparse representation. The fusion scheme not only is easy to use but also does not need to be artificially set weights. Moreover, it is consistent with the correlation between the classification error and the score obtained by the experimental analysis. In the field of face recognition, it has been shown that the two-step face recognition (TSFR) based on representation using the original training samples and the generated ''symmetric face'' training samples can achieve excellent face recognition performance. Face recognition based on multiplication fusion and TSFR proposed in this paper can further improve the recognition accuracy.

show abstract

Face recognition based on Volterra kernels direct discriminant analysis and effective feature classification

Cited by 12 publications

References 28 publications

Automatic Face Recognition Using Enhanced Firefly Optimization Algorithm and Deep Belief Network

Automatic Face Recognition Using Enhanced Firefly Optimization Algorithm and Deep Belief Network

Exact expectation analysis of the LMS adaptive identification of non‐linear systems

An Improved Two-Step Face Recognition Algorithm Based on Sparse Representation

Contact Info

Product

Resources

About